Scholarlink

Arrhythmia is a medical condition where a patient experiences an irregular heartbeat. The presence of arrhythmia indicates a much more serious heart condition, which, if not treated, may worsen and even lead to death. Hence, there is a requirement for a system that can detect the presence of such a condition. The patient’s ECG recordings were first converted into a spectrogram using short-time Fourier transform. These resultant spectrogram images were given as input to VGGNet models for arrhythmia classification. The VGG16 model exhibited 98.87% accuracy, 99.18% precision, 99.14% recall, and 97.94% specificity. On the other hand, VGG19 achieved 98.83% accuracy, 99.18% precision, 99.18% recall, and 97.95% specificity.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Deep Learning Approach for Arrhythmia Detection Using STFT-Based Spectrogram

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Shreya Chopra,
  • Nitigya Sambyal,
  • Anu Bajaj

摘要

Arrhythmia is a medical condition where a patient experiences an irregular heartbeat. The presence of arrhythmia indicates a much more serious heart condition, which, if not treated, may worsen and even lead to death. Hence, there is a requirement for a system that can detect the presence of such a condition. The patient’s ECG recordings were first converted into a spectrogram using short-time Fourier transform. These resultant spectrogram images were given as input to VGGNet models for arrhythmia classification. The VGG16 model exhibited 98.87% accuracy, 99.18% precision, 99.14% recall, and 97.94% specificity. On the other hand, VGG19 achieved 98.83% accuracy, 99.18% precision, 99.18% recall, and 97.95% specificity.